English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Segmentation and classification of cell nuclei in histopathology images usingdeep neural networks (DNNs) can save pathologists' time for diagnosing variousdiseases, including cancers, by automating cell counting and morphometricassessments. It is now well-known that the accuracy of DNNs increases with thesizes of annotated datasets available for training. Although multiple datasetsof histopathology images with nuclear annotations and class labels have beenmade publicly available, the set of class labels differ across these datasets.We propose a method to train DNNs for instance segmentation and classificationon multiple datasets where the set of classes across the datasets are relatedbut not the same. Specifically, our method is designed to utilize acoarse-to-fine class hierarchy, where the set of classes labeled and annotatedin a dataset can be at any level of the hierarchy, as long as the classes aremutually exclusive. Within a dataset, the set of classes need not even be atthe same level of the class hierarchy tree. Our results demonstrate thatsegmentation and classification metrics for the class set used by the testsplit of a dataset can improve by pre-training on another dataset that may evenhave a different set of classes due to the expansion of the training setenabled by our method. Furthermore, generalization to previously unseendatasets also improves by combining multiple other datasets with different setsof classes for training. The improvement is both qualitative and quantitative.The proposed method can be adapted for various loss functions, DNNarchitectures, and application domains.

Combining Datasets with Different Label Sets for Improved Nucleus Segmentation and Classification